REPRODUCTION. 21^ 



of wliicli is in every case a difficult and tedious work, teaclies that, 

 for example, tlie first primordia (protuberances) are acted upon hj 

 various mechanical influences which give rise to superposed whorls. 

 I have no cause to enter into a discussion of the conditions met 

 with in the flowers of Scrofularia^ because I do not know any 

 more about the origin of this genus than any one else does. 



Before entering into the discussion of the important relations of 

 the floral structure it is important to remember that the great 

 variety represented in the structure of flowers and fruits can no 

 longer be relegated to mere description, but must be considered 

 according to physiological adaptations. Although this change in 

 our science is comparatively recent, we are already enabled to give 

 physiological interpretations to many of the structural modifica- 

 tions ; always from a teleological standpoint. The purpose of many 

 of these modifications is to insure the most suitalile pollination and 

 fertilization. Under this category belong the following structural 

 modifications. 



A flower is said to be polysyminetrical or actinomorphio when 

 it may be symmetrically divided in at least two planes. (Slight 

 differences are not considered.) Illustration: the diagram (Fig. 

 139) of Liliuni. A flower is zygomorphic or syrmnetrical when 

 it can be divided symmetrically in only one plane. Sometimes 

 there is no plane of symmetry, when the flower is said to be 

 azyfjomorpJiic. The diagram of the labiate flower is an excel- 

 lent example of a zygomorphic flower (Fig. 142). The zygo- 

 morphic flowers are nearly alwaj's lateral. If a plant with zygo- 

 morphic flowers should huve a terminal flower, it is actinomorphie. 

 {Linaria vulgaris frequently shows this phenomenon. Such ter- 

 minal flowers are said to be jpeloric. ' ) 



The plane of symmetry of zygomorphic flowers usually, but 

 not always {Solanum, ^'Esculus)^ coincides with the median j)lane. 

 Tlie majority of flowers are therefore Tnedian-zygomorphic. The 

 median plane is that plane which bisects the axis of growth of a 

 lateral member as well as the axial member. Only a few flowers 

 are transversely zygomor j)liic ; that is, the plane of symmetry and 

 the median plane form an angle. In many orchid-flowers the 

 ovary or style rotates aljout its axis 180°, which brings the young 



' Lateral peloiic flowers are also reported. Whether such lateral positiou is real 

 or only apparent I am unable to stale. — Trans. 



